Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

He sat in the tiny plastic chair, watching his son, David, assemble a tower of blocks. A familiar figure approached, her smile kind. Ms. Peterson. He remembered her. She taught fourth grade when he was a student. The memory was a wound, fresh and sharp. He remembered the constant comparisons to his older brother, the feeling of never measuring up.

Now, she was looking at him, a quizzical expression on her face. "David's very bright," she said, her voice warm. "He reminds me of someone I used to know..."

The comment hung in the air, heavy with unspoken things. He couldn't meet her eyes. He felt a chill, a deep emptiness.

He forced a smile, but the words felt hollow on his tongue. He had failed his brother. Now, he felt he had failed his son too.

Emotion: guilty

Cluster: Shame / Guilt
PC1 (Valence): -1.62 Negative
PC2 (Disposition): 1.23

Role in Research

This story is one of 1,000 stories generated for the emotion guilty. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 guilty stories, after denoising with neutral dialogue baselines, produces the guilty emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the guilty vector is projected through the unembedding matrix.

Promoted:
S0.492
L0.398
탓0.354
been0.323
ashamed0.319
Suppressed:
de-0.400
(!)-0.279
la-0.276
😍-0.274
"!-0.264